MAPPING KNOWLEDGE: EVALUATING CONCEPT MAPS USING SOCIAL NETWORK ANALYSIS AND NETWORK SIMILARITY MEASURES

Traditional multiple-choice questions (MCQs) primarily assess factual recall and application but often fail to capture the depth and structure of student understanding. Addressing this limitation, the present study introduces a novel framework that leverages Social Network Analysis (SNA) and network similarity measures to evaluate concept maps generated from students' MCQ responses. The framework enhances assessment by identifying key concepts and assigning weighted importance based on degree centrality and influential nodes. By comparing student-constructed concept maps with an instructor’s reference map, it assesses the coherence and completeness of student knowledge. The study also evaluates learning depth through both self-constructed and quiz-derived concept maps, offering insights into students’ conceptual development. Furthermore, it clusters students based on performance, uncovers learning patterns, and identifies weaker concepts, unknown concepts, and misconceptions. This integrated approach facilitates efficient and consistent assessment, enables personalized instruction, and supports targeted pedagogical interventions, ultimately contributing to improved learning outcomes and deeper knowledge acquisition.